Claire Bowern's LINGUISTIC FIELDWORK leads young linguists through the methodological, ethical and financial necessities of successfully doing research among native speakers. There are chapters on what gear to take along, how to manage your notes, how to treat your native consultants, and how to ask for grants.
The book is mainly targeted to fieldworkers who seek to describe small, hitherto undocumented languages, the sort that have no written tradition yet and we may not even be sure of its basic phonology. In my field, Finno-Ugrian linguistics, most languages already have good basic descriptions and perhaps even native literature and other media, and fieldwork is more likely to consist of focusing on a very specific portion of the language instead of (my last trip, for example, was to hand out questionnaires on verbs of motion in Mari and Chuvash). Other fieldworkers are of the Labov variety, exploring linguistic differences among neighbourhoods and social classes in their own town. There's nothing specifically for these latter two types of fieldwork, but there's still much to learn from the general fieldwork advice that Bowern gives.
Indeed, the chapter on ethics here was pretty revelatory to me, as when I told my department I wished to do fieldwork, all it said was "That's nice" and gave me some money. Bowern, however, notes that much funding in the US is tied to application with a demonstrably ethical approach, and I had never seen some of these rules before.
I wish that Bowern had written more about how to deal with officialdom, especially when the state is an oppressive one with little care for the minority one goes to study. Completing obligatory registration procedures upon arrival in a community when everyone thinks you're a spy is difficult, and locals might not want to work with you if they fear they'll be punished for talking to foreigners. Finno-Ugrian and Turkic linguists sometimes encounter such irritations in Russia.
Also, the section on technology could have used some brushing up with outside help. The book is correct that recordings should not be kept in MP3 format, but its explanation for why not is incorrect. And Bowern wisely recommends keeping one's data in a clear, open format instead of a proprietary one like Microsoft Word, but she neglects to mention XML dialects, which have the best shot at longevity.
Still, even with some weaknesses, I imagine all linguists can get something from Bowern's book.